Deep Levels in High Resistivity AlGaN Films Grown by MOCVD
- PDF / 383,930 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 94 Downloads / 262 Views
[9]. EXPERIMENTAL Semi-insulating undoped AlxGal-xN films with compositions x=0.05, x=0. 15 and x=0.25 were grown on basal-face sapphire substrates by metallorganic chemical vapor deposition (MOCVD) using trimethylgallium, trimethylaluminum and ammonia as precursors. The growth was performed on low temperature AlGaN buffers, in a two-inlet vertical reactor, at 76 Torr, at 1100 °C. The thickness of the samples was about 2.5 rim. The experiments discussed below involved low temperature (90K) microcathodoluminescence (MCL) spectra measurements, the 239 Mat. Res. Soc. Symp. Proc. Vol. 512 © 1998 Materials Research Society
dark current versus temperature and the photocurrent versus temperature measurements (in the latter a deuterium UV lamp excitation was used) and the measurements of deep traps spectra using PICTS [10]. In the PICTS method a two-gate signal processing approach developed in DLTS is applied to photocurrent transients in semi-insulating samples with ohmic contacts. These measurements allow to determine the depth and capture cross section of the traps present in the material. As in DLTS, the temperature of the PICTS peak varies for varying time window settings tt/t 2 and this variation allows to determine the energies and capture cross sections of the traps producing the peaks [10]. PICTS spectra in this paper were taken with UV deuterium lamp excitation. A computer-actuated mechanical shutter was used to modulate the light. In addition to these measurements faster photocurrent transients (the system response time of 30 uts) were studied using an electronically triggered electron beam deviation system in a scanning electron microscope. Ohmic contacts for all such measurements were made by indium. The samples growth details and a detailed description of the experimental set-up can be found in [9]. RESULTS AND DISCUSSION The dark resistivity of the studied AlGaN samples was very high and meaningful dark current measurements could be made only for temperatures above room temperature. The temperature dependencies of the dark current in this temperature range showed that the Fermi level was pinned at respectively 0.5 eV, 0.35 eV and 0.6 eV in the samples with x=0.05, x=0.15 and x=0.25. The photocurrent was very high for photon energies higher than the bandgap while no measurable photosensitivity was detected for subbandgap excitation. Fig. 1 shows the temperature dependence of photocurrent for one of the samples (the x=0.25 sample) for UV excitation with deuterium lamp for three different excitation levels (attenuation of light was achieved by neutral filters; the results for the two other samples were qualitatively similar).
x=0.25, E,=0.35 eV 10"
0
'a
0
0110"1 10.10
10
150 150
10'I
200
250
200 250 Temperature (K)
Fig. 1. The temperature dependence of photocurrent in the x=0.25 sample.
3
eV
x=0.15, E=0.3 eV
10
300
I
6.0
6.5
7.0 7.5 1 1000/T (K" )
8.0
8.5
Fig. 2. Arrhenius plots used to determine the depth of the r-center in the three AIGaN samples.
240
It can be seen that the photo
Data Loading...
